- Landen: EU-27 en 3 EFTA-landen (IJsland, Noorwegen en Zwitserland, dus excl. Liechtenstein)
- Periodiciteit: kwartaalbasis
- Populatie: 15+
- Start: 1983
- NACE: 4 digit, rev. 1.1 van 1992 t.e.m. 2008
- ISCO: 2 of 3 digit, ISCO-88
- Disseminatie: Eurostat website, NuChronos dataset, Eurostat data requests
- 1.7 miljoen cases per kwartaal
- 0.2 tot 3.3% van de populatie
- Referentieweek gerandomiseerd over het kwartaal (sinds 2003)
Nadelen
country coverage
But: incomplete (not all use retrospective questions)
relatively comparable data
But: questions are not necessarily posed in the same way as in EU LFS
Recall problems increase as time goes by which can increase non-response and reduce quality of results
Retrospective questioning will be especially problematic in the case of proxy interviews
Coding practices can change over time
AND, of course…sample sizes
sample size:
approx. to 1,7 mio of individuals (2004, q2).
The sampling rates vary between 0.2% and 3.3% across the countries.
Cell sizes & reliability
Eurostat guidelines:minimum limits for cell sizes of quarterly results
(from 1500 to 21 000, depending on year, country)
Some solutions
Use of yearly averages (average of quarterly results)
Use of 3-year averages
Exclusion of diverging years / countries
Links
Website LFS: http://circa.europa.eu/irc/dsis/employment/info/data/eu_lfs/index.htm
Fiche LFS (EUROSTAT): http://epp.eurostat.ec.europa.eu/cache/ITY_SDDS/en/employ_esms.htm